Skip to main content
Open Analyze > Sentiment to review how detected brand mentions are described in monitored answers. Each analyzed mention receives a score on a 0 to 100 scale. Higher values represent more positive language in the analyzed context. The page also groups mentions into positive, neutral, and negative categories.

Choose the scope

Use the brand selector to analyze your brand, a competitor, or all your brands. The global provider, topic, tag, funnel, and date filters also apply. Confirm the selected brand before comparing sentiment with another page.

Overview tab

  • Brand Sentiment Score: the average score, its distribution into positive, neutral, and negative mentions, and the change against the previous period of the same length.
  • Sentiment Evolution: the score over time.
  • Sentiment Breakdown: the score, its change, and the number of scored mentions by provider, topic, tag, or funnel stage. Rows start with the largest decline, and a Low sample flag marks rows with few mentions.
  • Evidence Behind the Score: the Strongest Positive Mentions and the Priority Improvement Signals, with one example per answer and a link to open it.

Mentions tab

The mention list connects each score to its context: provider, score, prompt, justification, and date. Read the full answer when wording is ambiguous. A score summarizes detected tone; it does not establish customer opinion, brand reputation in the wider market, or the intent of the provider.

Compare periods

Keep the prompt set, models, countries, and brand relationship stable. A change in the mix of prompts can move the average even when similar answers receive similar scores. Treat changes on a small sample with caution.

Export

Use the export control for the current sentiment dataset. Record the filters and date range with any external report.